Effect of rifampin on enantioselective disposition and anti‐hypertensive effect of benidipine

Abstract

Aims

In vitro study showed that benidipine is exclusively metabolized by cytochrome P450 (CYP) 3A. This study evaluated the effect of rifampin on the enantioselective disposition and anti‐hypertensive effect of benidipine.

Methods

Benidipine (8 mg) was administered to healthy subjects with or without repeated rifampin dosing, in a crossover design. Plasma concentrations of (S)‐(S)‐(+)‐α and (R)‐(R)‐(−)‐α isomers of benidipine and blood pressure were measured for up to 24 h after dosing. In addition, CYP3A metabolic capacity was evaluated in each subject using oral clearance of midazolam.

Results

The exposure of (S)‐(S)‐(+)‐α‐benidipine was greater than that of (R)‐(R)‐(−)‐α‐benidipine by approximately three‐fold following single dose of benidipine. Repeated doses of rifampin significantly decreased the exposure of both isomers. Geometric mean ratios (GMRs) (95% CI) of C _max and AUC_∞ for (S)‐(S)‐(+)‐α‐benidipine were 0.14 (0.10–0.18) and 0.12 (0.08–0.18), respectively. GMRs (95% CI) of C _max and AUC_∞ for (R)‐(R)‐(−)‐α‐benidipine were 0.10 (0.06–0.17) and 0.10 (0.06–0.17), respectively. Oral clearances of both isomers were increased equally by approximately 10‐fold. There were no significant differences in cardiovascular effect following benidipine administration between control and rifampin treatment. CYP3A activity using midazolam did not appear to correlate with oral clearance of benidipine.

Conclusions

After single administration of racemic benidipine, enantioselective disposition of (S)‐(S)‐(+)‐α‐ and (R)‐(R)‐(−)‐α‐benidipine was observed. Treatments with rifampin significantly decreased the exposure of both isomers but appeared to marginally affect its blood pressure‐lowering effect in healthy subjects. Impact of coadministration of rifampin on the treatment effects of benidipine should be assessed in hypertensive patients.

What is Already Known about this Subject

• Benidipine is a racemic mixture of two isomers ((R)‐(R)‐(−)‐α and (S)‐(S)‐(+)‐α).

• Our previous study indicated that benidipine was metabolized by CYP3A4 and CYP3A5.

• Rifampin is a potent inducer of CYP3A4 that can decrease the antihypertensive effects of calcium channel blockers.

What this Study Adds

• A clinical study in healthy subjects was conducted to evaluate the enantioselective disposition of benidipine.

• This study describes for the first time the effect of rifampin on the pharmacokinetic and pharmacodynamic properties of benidipine and its enantiomers to establish the mechanism of the drug interaction.

Introduction

Benidipine is a dihydropyridine calcium‐channel blocker indicated for the treatment of systemic hypertension and angina pectoris. Unlike other calcium‐channel blockers, benidipine is a long‐lasting antihypertensive agent 1 and is synthesized for commercial use as a drug substance in a highly pure form 2. Moreover, benidipine is associated with a better prognostic effect than other calcium‐channel blockers in the treatment of patients with vasospastic angina 3. This effect can be attributed to the high affinity of benidipine for the cell membrane 4, 5 and slow dissociation from dihydropyridine binding sites 6, characteristics that are collectively described by the ‘membrane approach’ concept of benidipine 7. Benidipine exerts a renoprotective effect by inhibiting L‐, N‐ and T‐type calcium channels 8, 9, leading to reduced intraglomerular pressure 10 and plasma aldosterone levels 11, thereby reducing the urinary excretion of albumin 11, 12 and slowing the progression of nephropathy 13. Additionally, benidipine exerts a cardioprotective effect in myocardial ischaemia and reperfusion injury by increasing nitric oxide production, leading to coronary vasodilation and the attenuation of myocardial ischaemia progression 14.

Similar to most dihydropyridines, benidipine is a racemic mixture of two isomers ((R)‐(R)‐(−)‐α and (S)‐(S)‐(+)‐α 15, 16), in which the concentration of the (S)‐(S)‐(+)‐α isomer is almost two‐fold higher than that of the (R)‐(R)‐(−)‐α isomer 17. The (S)‐(S)‐(+)‐α isomer also exerts a 100‐fold more potent antihypertensive effect than the (R)‐(R)‐(−)‐α isomer 15. The previous investigation showed that the formation of metabolites in human liver microsomes was 2.9 and 2.4 μl min⁻¹ pmol⁻¹ P450 for (S)‐(S)‐(+)‐α and (R)‐(R)‐(−)‐α‐benidipine, respectively, indicating non‐stereoselectivity in the metabolism of benidipine 18.

In adults, benidipine is absorbed rapidly and reaches maximum plasma concentration (C _max) 0.5 and 0.75 h after dosing with 4 and 8 mg, respectively 19. Maximal cardiovascular effects were detected approximately 2 h after the administration of either dose. Benidipine is extensively metabolized in humans by CYP3A4 and CYP3A5 16, 18. Given the prominent role of CYP3A in benidipine metabolism, it is anticipated that the systemic exposure of benidipine and its enantiomeric disposition is significantly influenced by coadministration with potent CYP3A inducers. Rifampin has been shown to reduce the bioavailability of nifedipine 20, 21, most likely by mediating its metabolism in the gut wall 20. Other studies have also shown that coadministration of rifampin ameliorates the antihypertensive effects of calcium‐channel blockers including benidipine, amlodipine, nifedipine, and nisoldipine, leading to adverse events 22, 23, 24, 25. Enantioselectivity of a drug has been addressed in relation to pharmacokinetics, pharmacological activity and safety. This study aimed to evaluate the changes in enantioselective disposition of benidipine and its blood pressure‐lowering effect following multiple oral coadministration of rifampin.

Methods

Subjects

The study was approved by the Institutional Review Board at the Busan Paik Hospital, Inje University, Busan, South Korea, and was conducted at the Regional Clinical Trial Center in Busan Paik Hospital and the PharmacoGenomics Research Center of Inje University. Written informed consent was provided by each subject, and the study was conducted according to the Declaration of Helsinki as well as the International Conference on the Harmonization of the Technical Requirements for the Registration of Pharmaceuticals for Human Use – Good Clinical Practice (ICH‐GCP) guidelines. All subjects were assessed on the basis of medical history, physical examination and routine laboratory tests, including blood chemistry, haematology, urinalysis and 12‐lead electrocardiogram. All subjects were free from alcohol, herbal medicines, any drug other than study medicines, beverages containing caffeine, and grapefruit products for 7 days before and during study periods.

Study design

This was an open‐label, randomized, two‐way crossover study with two benidipine administration periods following an initial CYP3A phenotyping study (Figure 1). Participants were randomized into two groups to receive one of two treatment sequences and all received one oral dose of 2 mg of midazolam solution (Bukwang Pharm. Co., Ltd. Seoul, Korea) on Day 1 after overnight fasting. Blood samples (approximately 9 ml) were obtained from a forearm vein into a heparinized Vacutainer^® tube (Becton Dickinson, USA) prior to midazolam dosing and at 0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10 and 12 h after its administration. Samples were immediately centrifuged at 1000g for 10 min at 4°C and the resulting plasma was stored at −80°C until analysis. After a 3‐day wash‐out period, the first study period was initiated with the repeated oral administration of 600 mg rifampin (Yuhan Corporation, Seoul, Korea) once daily for 10 days (Day 5–14) to the participants in one of the groups. On Day 15, after overnight fasting, benidipine 8 mg (Yungjin Pharm. Co., Ltd., Seoul, Korea) was administered orally with 200 ml water after all subjects were confirmed to be in a healthy state by physical examination. Blood samples were drawn before and 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, 12 and 24 h after benidipine administration. Plasma was immediately obtained from the blood and stored at −80°C until analysis. After completion of the first study period and a 7‐day wash‐out period, rifampin was administered in the other group from Day 22 and single dose pharmacokinetic study of benidipine was repeated on all subjects on Day 32.

Figure 1.

Study design flow chart

For the assessment of benidipine pharmacodynamics, subjects were instructed to sit quietly while blood pressure and pulse rate were measured, and to rest for 10 min before beginning the blood pressure measurements. Systolic blood pressure, diastolic blood pressure and heart rate were recorded twice before dosing and at 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, 12 and 24 h after dosing. All blood pressure measurements were performed by the same research nurse using an automated device (AND UA‐767, A&D Company, Ltd, Tokyo, Japan).

Analytic methods

Plasma concentration of midazolam was determined by high‐performance liquid chromatography (HPLC)–tandem mass spectrometry (LC/MS/MS). Plasma samples (0.5 ml) were mixed with 10 μl of internal standard (IS; phenacetin, 1 μM) and extracted with 2 M NaOH (600 μl) and diethyl ether/methylene chloride (60:40 [vol./vol.], 5 ml). The dried residue was reconstituted in 200 μl of acetonitrile/water (30:70 [vol./vol.]) and transferred to an autosampler vial, and 10 μl was injected into the LC/MS/MS system. The HPLC system consisted of an Agilent 1100 series (Agilent Technology, Wilmington, DE, USA) and a Phenomenex Luna C18 (3 μm, 2 × 50 mm) reverse‐phase analytic column (Phenomenex, USA). The mobile phase, acetonitrile/0.1% formic acid (30:70 [vol./vol.]), was delivered at a total flow rate of 0.2 ml min⁻¹. Midazolam and IS concentration were determined with a PE SCIEX API 3000 LC/MS/MS system (Applied Biosystem, Canada) in electrospray ionization interface (positive ion) and multiple reaction‐monitoring modes. The mass transitions were 326 → 291 mass‐to‐charge ratio [m/z] for midazolam and 180 → 110 m/z for IS (collision energy, 35 eV). Standard curves were linear over the ranges of 0.5–20 ng ml⁻¹ for midazolam, and the intraday and interday coefficients of variation were less than 10%.

Plasma concentrations of benidipine were obtained using a previously described LC/MS/MS method developed and validated in our laboratory 17 with minor modifications. In brief, plasma samples (1.5 ml) with 20 μl of IS (deuterium, 100 ng ml⁻¹) were extracted with 5 N NaOH (1 ml) and diethyl ether (5 ml). Dried residues were resolved in 150 μl of mobile phase and 100 μl was injected into the LC/MS/MS system. Compounds were separated on a chiral stationary phase column (Chirobiotic V, 150 mm × 4.6 mm internal diameter, 5 μm particle size; Advanced Separation Technologies, Whippany, NJ, USA) with a mobile phase consisting of 10⁻²% acetic acid and 10⁻⁴% triethylamine in methanol. The protonated precursor ions and the related product ions of benidipine and IS were 506 → 174 m/z and 511 → 179 m/z, respectively (collision energy, 36 eV). The detection limit of (S)‐(S)‐(+)‐α‐ and (R)‐(R)‐(−)‐α isomer of benidipine was 0.002 ng ml⁻¹ (S/N > 5), and standard curves were linear over the range of 0.002–1 ng ml⁻¹. The intraday and interday coefficient of variation was less than 10%.

Data analysis

Pharmacokinetic analysis was carried out using WinNonlin^® (Pharsight Co., v4.0.1). A noncompartmental pharmacokinetic model with the single oral administration was used to calculate all parameters. The area under the concentration–time curve to the last observation time point (AUC_t) was calculated by the linear trapezoidal rule. AUC_∞ was calculated using the following equation: AUC_∞ = AUC_t + C _t/λz, where C _t is the last observed concentration and λz is the elimination rate constant. The terminal half‐life (t _1/2) was calculated as 0.693/λz. The oral clearance (CL/F) was calculated using an equation, dose/AUC_∞. Midazolam oral clearance was used as a phenotype index of CYP3A.

Maximum changes of blood pressure and pulse rate to their baseline values (baseline corrected‐means of maximum effect (∆E _max)) and time to E _max (t _Emax) were measured for pharmacodynamics analysis. The area under the effective (changes of blood pressure and pulse rate to the baseline–time curve from 0 to 24 h (∆AUEC)) was calculated.

Statistical analysis

The results are expressed as mean ± standard deviation (SD). The pharmacokinetic and pharmacodynamic parameters of benidipine with and without rifampin administration were compared using a paired t‐test. Geometric least squares mean ratios and their 95% confidence intervals for of area under the curve and maximum plasma concentrations for each drug were estimated. The relationship between midazolam oral clearance and the pharmacokinetic parameters of benidipine was assessed using Spearman's correlation coefficient (r _s). All data were analysed using SPSS for Windows v10.0.7 (SPSS, Inc., Chicago, IL, USA). Values of P < 0.05 were considered statistically significant.

Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY 26, and are permanently archived in the Concise Guide to PHARMACOLOGY 2017/18 27, 28, 29.

Results

Subjects

A total of 14 subjects were enrolled but one subject was excluded due to moderate diarrhoea during the rifampin administration period. Thus, 13 healthy male subjects (mean ± SD; aged 23.75 ± 2.06 years; weight 73.25 ± 7.3 kg; and height, 177.75 ± 3.8 cm) completed the entire study. At baseline, mean systolic and diastolic blood pressure of participants (mean ± SD; 125.1 ± 8.8 mmHg; 75.1 ± 6.75 mmHg for the period of rifampin coadministration and 121.3 ± 9.3 mmHg; 75.2 ± 8.8 mmHg for the period without the repeat of rifampin) appeared not statistically different between two study periods (P > 0.05).

Enantioselective disposition of benidipine

Following dosing with 8 mg racemic benidipine, plasma concentrations of (S)‐(S)‐(+)‐α and (R)‐(R)‐(−)‐α isomers of benidipine reached C _max at around 0.9 and 0.8 h, respectively, and declined in a bi‐phasic manner (Figure 2). (S)‐(S)‐(+)‐α‐benidipine showed significantly higher C _max (1.7 ng ml⁻¹ vs. 0.5 ng ml⁻¹, P < 0.0001) and longer elimination half‐life (5.2 vs. 2.8 h, P = 0.02) with significantly lower oral clearance (22.3 vs. 81.7 l h⁻¹ kg⁻¹, P < 0.0001) (Table 1). The AUC_∞ ratio between (S)‐(S)‐(+)‐α and (R)‐(R)‐(−)‐α isomers was approximately 3.5‐fold.

Figure 2.

Mean (± SD) concentration–time profile of benidipine after a single oral dose of 8 mg benidipine racemate with and without rifampin pre‐administration (n = 13)

Table 1.

Pharmacokinetic parameters of (S)‐(S)‐(+)‐α‐ and (R)‐(R)‐(−)‐α‐benidipine after a single oral dosing of 8 mg benidipine racemate in the absence or presence of rifampin in 13 healthy subjects

Parameter	Without rifampin (n = 13)	With rifampin (n = 13)	GMR (95%CI)	P‐valuea
(S)‐(S)‐(+)‐α‐ benidipine
C max (ng ml −1 )	1.7 ± 0.6	0.2 ± 0.1	0.14 (0.10–0.18)
AUC ∞ (ng h ml −1 )	3.2 ± 1.6	0.4 ± 0.3	0.12 (0.08–0.18)
t max (h)	0.9 ± 0.3	0.8 ± 0.5		0.67
t 1/2 (h)	5.2 ± 3.6	4.0 ± 3.2		0.40
CL/F (l h −1  kg −1 )	22.3 ± 12.4	206.9 ± 151.7	8.41 (5.19–11.64)	<0.001
(R)‐(R)‐(−)‐α ‐benidipine
C max (ng ml −1 )	0.5 ± 0.2c	0.1 ± 0.1	0.10 (0.06–0.17)
AUC ∞ (ng h ml −1 )	0.9 ± 0.4c	0.1 ± 0.2	0.10 (0.06–0.17)
t max (h)	0.8 ± 0.4	0.8 ± 0.5		0.82
t 1/2 (h)	2.8 ± 1.1b	2.2 ± 1.9		0.32
CL/F (l h −1  kg −1 )	81.7 ± 48.6c	1135.5 ± 1134.4	9.99 (2.80–17.18)	<0.001

AUC_∞, area under the plasma concentration–time curve from time zero to infinity; CL/F, oral clearance; C _max, maximum plasma concentration; GMR, Geometric mean ratio (95% confidence interval); t _max, time to reach C _max, t _1/2, terminal elimination half‐life

^aStudent t‐test

^b P < 0.05,

P < 0.01,

^c P < 0.001 compared with (S)‐(S)‐(+)‐α‐benidipine

The effect of rifampin on the pharmacokinetics of benidipine

Coadministration with repeated dose of 600 mg rifampin significantly decreased exposure of both benidipine enantiomers (Figure 2). Coadministration of rifampin led to an 86% (geometric mean ratio (95% confidence interval) [GMR (95% CI)]), 0.14 (0.10–0.18) and 90% (0.10 (0.06–0.17)) decrease in C _max of benidipine (S)‐(S)‐(+)‐α and (R)‐(R)‐(−)‐α isomers, respectively. The AUC_∞ of the (S)‐(S)‐(+)‐α‐ and (R)‐(R)‐(−)‐α isomers of benidipine decreased by 88% (0.12 (0.08–0.18)) and 90% (0.10 (0.06–0.17)), respectively (Table 1 and Figure 3). With rifampin coadministration, the CL/F increased by 8.41‐fold and 9.99‐fold of (S)‐(S)‐(+)‐α‐ and (R)‐(R)‐(−)‐α‐benidipine, respectively, which was not significantly different between the two isomers (P = 0.234). The increase in oral clearance of benidipine following multiple oral coadministration of rifampin appeared to vary so much among the subjects: it ranged from 1.4‐ to 29.2‐fold and from 1.6‐ to 51.9‐fold of benidipine (S)‐(S)‐(+)‐α and (R)‐(R)‐(−)‐α isomers, respectively. Time to maximum concentration (t _max) and t _1/2 of both benidipine enantiomers were not significantly affected by rifampin treatments (P > 0.05) (Table 1).

Figure 3.

Individual peak plasma concentration (C _max) (A), area under concentration–time curve (AUC_∞) (B), elimination half‐life (C), and oral clearance (CL/F) (D) values for benidipine (S)‐(S)‐(+)‐α isomers (solid symbols) and (R)‐(R)‐(−)‐α isomers (open symbols) after a single oral dose of 8 mg benidipine without and with repeated daily 600 mg rifampin administration (n = 13)

Midazolam oral clearance ranged from 0.57 l h⁻¹ kg⁻¹ to 2.6 l h⁻¹ kg⁻¹ (mean ± SD; 1.50 ± 0.69 l h⁻¹ kg⁻¹). Pharmacokinetic parameters of each benidipine isomer did not appear to correlate with midazolam oral clearance, with or without rifampin coadministration (data not shown).

Pharmacodynamics of benidipine

The effect of rifampin on the cardiovascular responses following oral administration of benidipine is summarized in Table 2. Following single dosing of 8 mg benidipine, systolic and diastolic blood pressure significantly decreased in both periods without and with rifampin coadministration: ΔE _max for systolic blood pressure without and with rifampin coadministration was 10.4 ± 6.1 and 12.5 ± 3.4 mmHg, respectively, and ΔE _max for diastolic blood pressure was 15.2 ± 10.5 and 16.7 ± 8.1 mmHg, respectively. ΔE _max for heart rate was 14.3 ± 6.9 and 16.6 ± 9.5 beats min⁻¹, respectively. The maximum change values of blood pressure and pulse rate were not significantly different between with and without rifampin coadministration (P > 0.05). In addition, AUEC for the changes in blood pressure and heart rate did not appear to show any significant difference in relation to rifampin administration (P > 0.05) (Table 2).

Table 2.

Changes of blood pressure and heart rates following a single oral dose of 8 mg benidipine racemate in the absence or presence of rifampin in 13 healthy subjects

Parameters	Without rifampin	With rifampin	P‐valuea
Systolic blood pressure
Baseline (mmHg)	121.3 ± 9.3	125.1 ± 8.8	0.31
ΔE max (mmHg)	−10.4 ± 6.1	−12.5 ± 3.4	0.59
t Emax (h)	4.7 ± 5.75	2.35 ± 2.7	0.09
ΔAUEC 0–24 (mmHg h)	22.2 ± 230.05	−18.3 ± 85.45	0.39
Diastolic blood pressure
Baseline (mmHg)	75.2 ± 8.8	75.1 ± 6.75	0.97
ΔE max (mmHg)	−15.2 ± 10.5	−16.7 ± 8.1	0.75
t Emax (h)	3.4 ± 2.9	5.8 ± 2.65	0.18
ΔAUEC 0–24 (mmHg h)	−125.8 ± 204.9	−129.85 ± 109.3	0.47
Heart rate
Baseline (beat min −1 )	68.9 ± 7.2	71.2 ± 8.1	0.47
ΔE max (beat min −1 )	16.6 ± 9.5	14.3 ± 6.9	0.26
t Emax (h)	6.35 ± 4.8	11.6 ± 6.7	0.28
ΔAUEC 0–24 (beat h min −1 )	159.1 ± 200.1	145.6 ± 160.3	0.84

ΔE_max, baseline corrected‐means of maximum effect, T_Emax,time to E_max, and ΔAUEC_0–24,baseline corrected‐area under the effect‐time curve from time 0 to 24 h

^aStudent t‐test

Discussion

This study evaluated the enantioselective disposition of benidipine in healthy Korean subjects, as well as its pharmacokinetic and pharmacodynamic properties in relation to repeated rifampin administration and phenotypic CYP3A activity using midazolam. Benidipine, a dihydropyridine‐derived calcium‐channel blocker, is composed of two optical isomers and is characterized by the slow onset of action and long‐acting effect, high lipid solubility and vascular selectivity 14. The (S)‐(S)‐(+)‐α isomer of benidipine shows higher receptor binding affinity and, subsequently, a more potent effect than the (R)‐(R)‐(−)‐α isomer 5. In this study, plasma concentrations of the (S)‐(S)‐(+)‐α isomer after single oral administration of racemic benidipine appeared to be consistently higher than those of the (R)‐(R)‐(−)‐α isomer in all subjects, which is consistent with the previous investigation showing approximately two‐fold higher C _max and 1.9‐fold higher AUC_∞ for the (S)‐(S)‐(+)‐α isomer than the (R)‐(R)‐(−)‐α isomer 30.

Benidipine is rapidly absorbed and extensively metabolized after oral administration 16, resulting in a rapid decrease in plasma concentration of the parent drug. The elimination half‐life of benidipine reported in the previous studies ranged from 1 to 5.3 h 16, 30, 31, 32. The analytical method used in the present study had high enough sensitivity (detection limit of benidipine isomer, 0.002 ng ml⁻¹) that it could determine the concentration of each isomer in plasma samples until 24 h after benidipine dosing. Peak benidipine concentration was achieved at 0.5–1 h after drug administration and declined bi‐exponentially, showing a terminal half‐life of 5.2 ± 3.6 and 2.8 ± 1.1 h for the (S)‐(S)‐(+)‐α and (R)‐(R)‐(−)‐α isomer, respectively. These results are similar to those observed in the studies performed in Caucasian 16 and Japanese volunteers 32. The characteristically high receptor‐binding capacity and lipid solubility of benidipine suggest that it is rapidly disposed into the effect compartment and slowly eliminated from it, which lead to a prolonged pharmacological action being maintained in the cardiovascular system. The relatively lower C _max and shorter t _1/2 of the (R)‐(R)‐(−)‐α isomer than those of the (S)‐(S)‐(+)‐α isomer suggest that both presystemic and systemic metabolism are involved in the enantioselective disposition of benidipine.

The in vitro study revealed that benidipine is mainly metabolized by CYP3A4 and CYP3A5 18. CYP3A enzymes are found abundantly in the human liver and intestine, playing a major role in first‐pass metabolism. Given that benidipine is a substrate of CYP3A, a correlation between CYP3A activity and the pharmacokinetics of benidipine is anticipated. However, the current study showed that there was no significant correlation between midazolam oral clearance and benidipine oral clearance (r _s = −0.324 and −0.137; P = 0.28 and 0.66, for (S)‐(S)‐(+)‐α‐ and (R)‐(R)‐(−)‐α isomer, respectively). These results suggest that other factor(s) than CYP3A metabolic capacity may contribute to interindividual variability in oral bioavailability of benidipine. Erythromycin breath testing, used as a probe of CYP3A activity and a substrate of P‐glycoprotein (P‐gp), did not correlate with midazolam clearance, which is likely to be attributed to the potential confounding factor of individual P‐gp activity 33, 34, 35, 36. Although benidipine has not been clearly elucidated to be a substrate of P‐gp, intestinal P‐gp activity may contribute to the disposition of benidipine and its interindividual variability. A large number of substrates of CYP3A appear to also be substrates of P‐gp. Benidipine shows a potent inhibitory effect on P‐gp‐mediated digoxin transport 37, implying that it may also be a substrate of P‐gp. In addition, benidipine is highly lipid‐soluble 7, a common characteristic of P‐gp substrates. Further study is required to assess the involvement of P‐gp in the disposition of benidipine and its relative contribution.

Repeated daily dosing of 600 mg rifampin significantly decreased the plasma concentration of benidipine. Rifampin decreased AUC_∞ and C _max of benidipine by over 86% and increased its oral clearance by more than 10‐fold compared to values observed in the benidipine only period. However, no significant difference in t _1/2 of benidipine was observed between with or without rifampin treatment, indicating that rifampin coadministration primarily affects the bioavailability of benidipine and is unlikely to significantly decrease its systemic clearance. Similar findings were observed in the effects of rifampin on the pharmacokinetics of nifedipine and verapamil; while rifampin did not affect their clearance when given intravenously, it increased oral clearance of both drugs and thereby reduced their bioavailability 20, 38. Rifampin has been known to exert the greatest effects on the pharmacokinetics of drugs administered orally with extensive first‐pass metabolism, including midazolam and simvastatin 3, 39. As a potent inducer of CYP3A4, rifampin induces both intestinal and liver CYP3A4 enzymes. However, from the collective findings of previous rifampin interaction studies and the present study, induction of pre‐hepatic gut wall metabolism may play a pivotal role in the bioavailability of CYP3A‐substrate drugs. In addition, given that P‐gp is possibly involved in the disposition of benidipine, a differential effect of rifampin treatments on the intestinal tract and hepatic tissues may also contribute to the currently observed finding. Given the timing of administration of the substrate and perpetrator drugs, it is complex to assess the drug–drug interaction (DDI) potential of investigational drug. The magnitude of DDI depends primarily on the duration and timing of rifampicin administration; however, this notion is still controversial. A 10‐day treatment with rifampin 600 mg daily increased CYP3A4 expression in human enterocytes by 288% 40. In a study conducted in humans, 5‐day treatment with rifampin 600 mg daily decreased the AUC of oral midazolam by approximately 96–98% 3, 41. In addition, following a 6‐day treatment with rifampicin, AUC of nilvadipine was reduced by 96% 2, whereas after 7 days of rifampin treatment (600 mg day⁻¹), oral clearance of nifedipine increased from 1.5 l min⁻¹ to 20.9 l min⁻¹ (P < 0.01) 20. In the case of racemic verapamil, the maximal inductive effect on S‐ and R‐verapamil trough concentrations of rifampin was observed from the 8th to the 12th day of treatment (600 mg day⁻¹) 38. Therefore, to minimize the time and resource needed and to observe the effect of CYP3A inducer on the enantioselective disposition of benidipine, rifampin was dosed for 10 days in our present study.

Benidipine has beneficial effects in the treatment of hypertension and angina pectoris compared with other calcium‐channel blocking agents. In the present study, the cardiovascular effect of benidipine after a single 8 mg administration in healthy volunteers showed that benidipine reduced diastolic blood pressure to a greater extent than systolic pressure. Compared with baseline values, diastolic blood pressure decreased significantly until 24 h after dosing, with the mean maximum decrease (ΔE _max) of 15.2 ± 10.5 mmHg at 3.4 h after dosing. These findings conform to the previously reported characteristics of benidipine following single‐dose administration 19. Heart rate tended to increase accordingly, but only slightly changed (mean increase of 6.35 beats min⁻¹, P > 0.05). Benidipine has been known for its potent blood pressure‐reducing effect while maintaining the heart rate 7, 14, 42. Unlike most calcium‐channel antagonists, which induce tachycardia via neurohumoral activation, benidipine causes only a minimal elevation in plasma catecholamine level 14. The magnitude of the decrease in blood pressure was markedly smaller than in the previously studied hypertensive patients 43, suggesting a reduced pharmacological effect in normotensive subjects. One investigation showed that the decrease in mean blood pressure was almost 20 mmHg at 24 h after the single oral administration of 8 mg benidipine in patients with essential hypertension 7, 43.

Decreased drug concentrations of several calcium‐channel antagonists followed by rifampin treatments reduce their therapeutic effects 23, 24, 44, 45, 46. The effect of rifampin on the blood pressure‐lowering effect of benidipine was not significant in the present study, although plasma concentrations were significantly reduced. It is noted that a single dose of benidipine is insufficient to explore the effect of rifampin on its pharmacodynamics. In addition, this study was performed in healthy normotensive subjects. A study in hypertension patients following long‐term treatment of both drugs may provide an insight into treatment failure and a decrease in other beneficial effects of benidipine. In the previous evaluation of the effects of rifampin on the pharmacokinetics and pharmacodynamics of nilvadipine in healthy subjects, no significant changes in the pharmacodynamic response of nilvadipine were observed following coadministration with rifampin 2.

In conclusion, benidipine displays an enantiomerselective disposition in healthy Korean subjects, showing higher concentrations of the (S)‐(S)‐(+)‐α isomer than of the (R)‐(R)‐(−)‐α isomer. There was no association between phenotypic CYP3A activity using oral midazolam and oral clearance of benidipine. This may imply that the effect of rifampin on the pharmacokinetics of benidipine is mainly mediated by a profound change in disposition factor(s) other than CYP3A4, e.g. P‐gp. Rifampin markedly decreased the exposure of benidipine through a nonstereoselective induction potential. Repeated treatments of rifampin showed negligible effect on the blood pressure‐lowering effect following single oral dose administration of 8 mg benidipine in healthy subjects. Given the significant effect of rifampin treatment on the disposition of benidipine, clinical benefits of benidipine should be assessed when rifampin is coadministered in hypertensive patients whose blood pressure is controlled by benidipine.

Competing Interests

There are no competing interests to declare.

This study was supported by a grant from the Korea Health Industry Development Institute, Ministry of Health & Welfare, Republic of Korea (grant number of HI15C1537). We thank all clinical trial staff members for their contributions to this study. J.S. is currently employed by the US Food and Drug Administration. His contribution to the manuscript was based on his prior employment, and the current manuscript does not necessarily reflect any position of the US Food and Drug Administration or the US government.

Sunwoo Y. E., Nguyen P. T. T., Chien C. M., Ryu J. Y., Shon J., and Shin J.‐G. (2019) Effect of rifampin on enantioselective disposition and anti‐hypertensive effect of benidipine, Br J Clin Pharmacol, 85, 737–745. 10.1111/bcp.13848.